# Importing the libraries to Jupyter Notebook
import pandas as pd 
import numpy as np
import matplotlib.pyplot as plt 
import seaborn as sns
import geopandas as gpd

# Step 2 Read CSV file
# Load csv file to pandas dataframe =
World_Population = pd.read_csv("C:/Users/brukt/Downloads/world-population.csv")

World_Population

# Step 3 Explore the data
# Checking the first 5 rows from the data frame to understand the structure
World_Population.head()

# Get basic info about the database such as numbers of non-nullvalue, column names, data type
print(World_Population.info())

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 3618 entries, 0 to 3617
Data columns (total 14 columns):
 #   Column                         Non-Null Count  Dtype  
---  ------                         --------------  -----  
 0   country                        3618 non-null   object 
 1   Year                           3618 non-null   int64  
 2   Population                     3618 non-null   int64  
 3   Yearly %   Change              3618 non-null   object 
 4   Yearly  Change                 3618 non-null   int64  
 5   Migrants (net)                 3618 non-null   float64
 6   Median Age                     3618 non-null   float64
 7   Fertility Rate                 3618 non-null   float64
 8   Density (P/Km²)                3618 non-null   int64  
 9   Urban Pop %                    3618 non-null   object 
 10  Urban Population               3618 non-null   object 
 11  Country's Share of  World Pop  3618 non-null   object 
 12  World Population               3618 non-null   int64  
 13  Rank                           3618 non-null   int64  
dtypes: float64(3), int64(6), object(5)
memory usage: 395.8+ KB
None

# View summary statistics for numerical columns
World_Population.describe()

# Step 4 cleaning datas
# Identifying missing values
World_Population.isnull().sum()

country                          0
Year                             0
Population                       0
Yearly %   Change                0
Yearly  Change                   0
Migrants (net)                   0
Median Age                       0
Fertility Rate                   0
Density (P/Km²)                  0
Urban Pop %                      0
Urban Population                 0
Country's Share of  World Pop    0
World Population                 0
Rank                             0
dtype: int64

# check if there is duplicated rows
World_Population.duplicated().sum()

0

## Cleaning columns with exta space 

## displaying the current column
print(World_Population.columns)

Index(['country', 'Year', 'Population', 'Yearly %   Change', 'Yearly  Change',
       'Migrants (net)', 'Median Age', 'Fertility Rate', 'Density (P/Km²)',
       'Urban Pop %', 'Urban Population', 'Country's Share of  World Pop',
       'World Population', 'Rank'],
      dtype='object')

## replace the original columns by removing extra sppaces
## (1) we can either do it one by one by using this code
World_Population = World_Population.rename(columns={'Urban  Pop %': 'Urban Pop %'})
## display the updated column names
print(World_Population.columns)

## (2) we can remove all extra space from all columns in the World_Population by using "str.strip" 
World_Population.columns = World_Population.columns.str.strip()

Index(['country', 'Year', 'Population', 'Yearly %   Change', 'Yearly  Change',
       'Migrants (net)', 'Median Age', 'Fertility Rate', 'Density (P/Km²)',
       'Urban Pop %', 'Urban Population', 'Country's Share of  World Pop',
       'World Population', 'Rank'],
      dtype='object')

World_Population

World_Population.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 3618 entries, 0 to 3617
Data columns (total 14 columns):
 #   Column                         Non-Null Count  Dtype  
---  ------                         --------------  -----  
 0   country                        3618 non-null   object 
 1   Year                           3618 non-null   int64  
 2   Population                     3618 non-null   int64  
 3   Yearly %   Change              3618 non-null   object 
 4   Yearly  Change                 3618 non-null   int64  
 5   Migrants (net)                 3618 non-null   float64
 6   Median Age                     3618 non-null   float64
 7   Fertility Rate                 3618 non-null   float64
 8   Density (P/Km²)                3618 non-null   int64  
 9   Urban Pop %                    3618 non-null   object 
 10  Urban Population               3618 non-null   object 
 11  Country's Share of  World Pop  3618 non-null   object 
 12  World Population               3618 non-null   int64  
 13  Rank                           3618 non-null   int64  
dtypes: float64(3), int64(6), object(5)
memory usage: 395.8+ KB

# Handling missing values 
# drop rows with missing values(if any)
World_Population.dropna(inplace=True)

World_Population.isnull().sum()

country                          0
Year                             0
Population                       0
Yearly %   Change                0
Yearly  Change                   0
Migrants (net)                   0
Median Age                       0
Fertility Rate                   0
Density (P/Km²)                  0
Urban Pop %                      0
Urban Population                 0
Country's Share of  World Pop    0
World Population                 0
Rank                             0
dtype: int64

# Total population for the latest year 
# First define the latest year, then the total population of the latest year
latest_year = World_Population['Year'].max()
total_population_latest_year = World_Population[World_Population['Year'] == latest_year]['Population'].sum()
print(f'The World Population in {latest_year}: {total_population_latest_year}')

The World Population in 2020: 7794099349

latest_year = World_Population['Year'].max()
total_population_latest_year = World_Population[World_Population['Year'] == latest_year]['Population'].sum()
print(f"Total World Population in {latest_year}: {total_population_latest_year}")

Total World Population in 2020: 7794099349

# formatting the numbers with comma in order to separate them 
total_population_formatted = f"{total_population_latest_year:,}"
print(f'The World population in {latest_year}: {total_population_formatted}')

The World population in 2020: 7,794,099,349

latest_year = World_Population['Year'].max()

population_by_country_latest_year = World_Population[World_Population['Year'] == latest_year].groupby('country')['Population'].sum()
population_by_country_latest_year.head(10)

country
Afghanistan            38928346
Albania                 2877797
Algeria                43851044
Angola                 32866272
Antigua And Barbuda       97929
Argentina              45195774
Armenia                 2963243
Aruba                    106766
Australia              25499884
Austria                 9006398
Name: Population, dtype: int64

## latest year population by country(grouping)
population_by_country_latest_year = World_Population[World_Population['Year'] == latest_year].groupby('country')['Population'].sum().sort_values(ascending=False)
population_by_country_latest_year.head(10)

country
China            1439323776
India            1380004385
United States     331002651
Indonesia         273523615
Pakistan          220892340
Brazil            212559417
Nigeria           206139589
Bangladesh        164689383
Russia            145934462
Mexico            128932753
Name: Population, dtype: int64

## Average Population for latest year by country
Average_population_latest_year = World_Population[World_Population['Year'] == latest_year]['Population'].mean()
print(f'The Average Population in {latest_year}: {Average_population_latest_year}')

The Average Population in 2020: 38776613.676616915

## Data visualisation
# Population distribution for latest year
plt.figure(figsize=(10,6))
sns.histplot(World_Population[World_Population['Year'] == latest_year]['Population'], bins=30, kde=True)
plt.title(f'Population Distribution in {latest_year}')
plt.xlabel('Population')
plt.ylabel('Frequency')
plt.show()

C:\ProgramData\anaconda3\Lib\site-packages\seaborn\_oldcore.py:1119: FutureWarning: use_inf_as_na option is deprecated and will be removed in a future version. Convert inf values to NaN before operating instead.
  with pd.option_context('mode.use_inf_as_na', True):

top_countries = World_Population[World_Population['Year'] == 2020].nlargest(5, 'Population')

plt.figure(figsize=(12, 6))
sns.barplot(data=top_countries, x='Population', y='country')
plt.title('Top 5 Most Populous Countries in 2020')
plt.xlabel('Population')
plt.ylabel('Country')
plt.show()

# TOP 10 most populated countries byy the latest year
population_by_country_latest_year = World_Population[World_Population['Year'] == latest_year].groupby('country')['Population'].sum().sort_values(ascending=False)
Top_10_countries_latest_year = population_by_country_latest_year.head(10)

plt.figure(figsize=(10,6))
sns.barplot(x=Top_10_countries_latest_year.values, y=Top_10_countries_latest_year.index, palette='viridis')
plt.title(f'Top 10 Most Populated countries by {latest_year}')
plt.xlabel('Population')
plt.ylabel('country')
plt.show()

import plotly.express as px

# Interactive plot with plotly
fig = px.line(World_Population, x='Year', y='Population', color='country', title='Population Growth Over Time')
fig.show()

#### Population growth analysis
# Plot population growth for a specific country 
country = 'United Kingdom'
country_data = World_Population[World_Population['country'] == country]

plt.figure(figsize=(10,6))
plt.plot(country_data['Year'],country_data['Population'], marker = 'o')
plt.title(f'Population Growth in {country}')
plt.xlabel('Year')
plt.ylabel('Population')
plt.grid(True)
plt.show()

# plotting the median age trend for a specific cocountry
# median age trend of the united kingdon
country = 'United Kingdom'
country_data = World_Population[World_Population['country'] == country]
plt.figure(figsize=(12,6))
plt.plot(country_data['Year'], country_data['Population'], marker = "o", color = 'green', linestyle='-', linewidth=0.5, alpha= 0.7)
plt.title(f'The Median Age in {country}')
plt.xlabel('Year')
plt.ylabel('Median Age')
plt.grid(True)
plt.show()

## Population growth ananlysis
# camparative analysis betweeen united kingdom and united france
# first we determine the countries
country1 = 'France'
country2 = 'United Kingdom'
# 2nd determining the population of each country
country1_data = World_Population[World_Population['country'] == country1]
country2_data = World_Population[World_Population['country'] == country2]
# 3rd ploting the analysis using graph
plt.figure(figsize=(10,6))
plt.plot(country1_data['Year'], country1_data['Population'], marker = '*', label = country1, color = 'purple')
plt.plot(country2_data['Year'], country2_data['Population'], marker = 'o', label = country2, color = 'orange')
plt.title('The Population Growth comparison')
plt.xlabel('Year')
plt.ylabel('Population')
plt.legend()
plt.grid(True)
plt.show()

import plotly.express as px

# Ploting an urban population percentage trend for a specific country
plt.figure(figsize=(10,6))
plt.plot(country_data['Year'], country_data['Urban Pop %'], marker = 'o', color = 'violet')
plt.title(f'Urban Population Percentage in {country}')
plt.xlabel('Year')
plt.ylabel('Urban pop %')
plt.grid(True)
plt.show()

# correlation analysis : These insights help in understanding the demographic and population dynamics across different countries over time.
# 1 st Filter out non-numeric columns because the world population data frame consists not only numbers but floats,...
numeric_df = World_Population.select_dtypes(include=[float, int])

correlation_matrix= numeric_df.corr()

sns.heatmap(correlation_matrix, annot=True, cmap='coolwarm')
plt.title('Correlation Matrix')
plt.show()

#### the correlation int this data heatmap shows that the relation ship between different numerical variables ranging from -1 to 1 indicates:

# (1) Year: Positively correlated with Population (0.075), Yearly Change (0.03), Density (P/Km²) (0.07), and World Population (1).
#        Negatively correlated with Median Age (-0.53) and Rank (-0.0012).
#    This suggests that as time progresses, the world population and density increase, while median age decreases.
# (2) Population: Positively correlated with Year (0.075), Yearly Change (0.31), Density (P/Km²) (0.41), and World Population (1).
#                   Negatively correlated with Rank (-0.53).
#     This implies that countries with higher populations tend to have a higher yearly change and density but a lower rank.
# (3)  Yearly change: Positively correlated with Population (0.31) and Density (P/Km²) (0.15).
#    This indicates that countries with a higher population tend to have a higher yearly change in population.
# (4) Migrants rate: No strong correlations with other variables.
#    This suggests that net migration does not significantly impact other variables in this dataset.
# (5) Median age: Negatively correlated with Population (-0.53) and Yearly Change (-0.5).
#                  Positively correlated with Fertility Rate (0.31).
#     This indicates that countries with a higher median age tend to have lower populations and yearly changes but higher fertility rates.
# (6) Fertility Rate: Positively correlated with Median Age (0.31).
#    This suggests that higher fertility rates are associated with a higher median age in the population.
# (7) Density (p/km^2) : Positively correlated with Population (0.41) and Yearly Change (0.15).
#    This implies that higher population density is associated with higher populations and yearly changes.
# (8) Word population: Perfectly correlated with Year (1), indicating that world population data changes consistently with the year.
#   Positively correlated with Population and Density.
# (9) Rank: Negatively correlated with Population (-0.53).
# This suggests that countries with a higher population have a lower rank (where a lower rank number indicates a higher population rank).
# conclussion: 
# Population Growth: As the year progresses, the world population grows, which is expected.
# Population Density: Higher population countries tend to have higher population densities.
# Median Age: Higher populations and yearly changes are associated with a lower median age, suggesting that younger populations are growing faster.
# Fertility Rate: There is a complex relationship where higher fertility rates are found in populations with a higher median age.

World_Population.to_csv("C:/Users/brukt/Downloads/world-population.csv", index=False)

World_Population = pd.read_csv("C:/Users/brukt/Downloads/world-population.csv")

World_Population

World_Population.info()

<class 'pandas.core.frame.DataFrame'>
RangeIndex: 3618 entries, 0 to 3617
Data columns (total 14 columns):
 #   Column                         Non-Null Count  Dtype  
---  ------                         --------------  -----  
 0   country                        3618 non-null   object 
 1   Year                           3618 non-null   int64  
 2   Population                     3618 non-null   int64  
 3   Yearly %   Change              3618 non-null   object 
 4   Yearly  Change                 3618 non-null   int64  
 5   Migrants (net)                 3618 non-null   float64
 6   Median Age                     3618 non-null   float64
 7   Fertility Rate                 3618 non-null   float64
 8   Density (P/Km²)                3618 non-null   int64  
 9   Urban Pop %                    3618 non-null   object 
 10  Urban Population               3618 non-null   object 
 11  Country's Share of  World Pop  3618 non-null   object 
 12  World Population               3618 non-null   int64  
 13  Rank                           3618 non-null   int64  
dtypes: float64(3), int64(6), object(5)
memory usage: 395.8+ KB

from geopy.geocoders import Nominatim 
from IPython.display import IFrame, display
import time
import folium

# filtering the world data for the year 2020
world_pop_2020 = World_Population[World_Population['Year'] == 2020]

geolocator = Nominatim(user_agent="geoapiWorld")

def get_coordinates(country):
    try:
        location = geolocator.geocode(country)
        return (location.latitude, location.longitude)
    except:
        return (None, None)

# Get coordinates for each country for the year 2020
world_pop_2020['coordinates'] = world_pop_2020['country'].apply(get_coordinates)

# Drop rows where coordinates could not be found
world_pop_2020 = world_pop_2020.dropna(subset=['coordinates'])

# Separate latitude and longitude into separate columns
world_pop_2020['latitude'] = world_pop_2020['coordinates'].apply(lambda x: x[0])
world_pop_2020['longitude'] = world_pop_2020['coordinates'].apply(lambda x: x[1])

C:\Users\brukt\AppData\Local\Temp\ipykernel_11812\1735024007.py:11: SettingWithCopyWarning:


A value is trying to be set on a copy of a slice from a DataFrame.
Try using .loc[row_indexer,col_indexer] = value instead

See the caveats in the documentation: https://pandas.pydata.org/pandas-docs/stable/user_guide/indexing.html#returning-a-view-versus-a-copy

# Create a map centered around the mean latitude and longitude
# to change the map style use 'tiles='OpenStreetMap','Stamen Terrain','Stamen Toner',
#'Stamen Watercolor','CartoDB positron','CartoDB dark_matter'
m = folium.Map(
    location=[world_pop_2020['latitude'].mean(), world_pop_2020['longitude'].mean()], 
    zoom_start=2, 
    tiles='Openstreetmap'
)

# Add points to the map
for _, row in world_pop_2020.iterrows():
    folium.CircleMarker(
        location=[row['latitude'], row['longitude']],
        radius=5,
        popup=(
            f"Country: {row['country']}<br>"
            f"Population: {row['Population']}<br>"
            f"Yearly % Change: {row['Yearly %   Change']}<br>"
            f"Median Age: {row['Median Age']}<br>"
            f"Fertility Rate: {row['Fertility Rate']}<br>"
            f"Density (P/Km²): {row['Density (P/Km²)']}<br>"
            f"Urban Pop %: {row['Urban Pop %']}<br>"
            f"Urban Population: {row['Urban Population']}<br>"
        ),
        color='blue',
        fill=True,
        fill_color='blue'
    ).add_to(m)

# Save the map to an HTML file
m.save('map.html')

# to display the map with in the notebook
display(IFrame('map.html', width=700, height=500))

	country	Year	Population	Yearly % Change	Yearly Change	Migrants (net)	Median Age	Fertility Rate	Density (P/Km²)	Urban Pop %	Urban Population	Country's Share of World Pop	World Population	Rank
0	Afghanistan	2020	38928346	2.33 %	886592	-62920.0	18.4	4.56	60	25.4 %	9904337	0.50 %	7794798739	37
1	Afghanistan	2019	38041754	2.34 %	869833	-62920.0	17.4	5.26	58	25.2 %	9582625	0.49 %	7713468100	37
2	Afghanistan	2018	37171921	2.41 %	875808	-62920.0	17.4	5.26	57	24.9 %	9273302	0.49 %	7631091040	38
3	Afghanistan	2017	36296113	2.58 %	913081	-62920.0	17.4	5.26	56	24.7 %	8971472	0.48 %	7547858925	39
4	Afghanistan	2016	35383032	2.82 %	969429	-62920.0	17.4	5.26	54	24.5 %	8670939	0.47 %	7464022049	39
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
3613	Zimbabwe	1975	6293875	3.54 %	200914	-9109.0	15.4	7.40	16	19.3 %	1215331	0.15 %	4079480606	79
3614	Zimbabwe	1970	5289303	3.42 %	163625	-8400.0	15.6	7.40	14	17.0 %	898584	0.14 %	3700437046	79
3615	Zimbabwe	1965	4471177	3.43 %	138899	-3002.0	16.0	7.30	12	14.4 %	644767	0.13 %	3339583597	91
3616	Zimbabwe	1960	3776681	3.28 %	112679	-1501.0	17.2	7.00	10	12.5 %	472478	0.12 %	3034949748	87
3617	Zimbabwe	1955	3213286	3.19 %	93287	-901.0	18.1	6.80	8	11.5 %	371106	0.12 %	2773019936	91

	country	Year	Population	Yearly % Change	Yearly Change	Migrants (net)	Median Age	Fertility Rate	Density (P/Km²)	Urban Pop %	Urban Population	Country's Share of World Pop	World Population	Rank
0	Afghanistan	2020	38928346	2.33 %	886592	-62920.0	18.4	4.56	60	25.4 %	9904337	0.50 %	7794798739	37
1	Afghanistan	2019	38041754	2.34 %	869833	-62920.0	17.4	5.26	58	25.2 %	9582625	0.49 %	7713468100	37
2	Afghanistan	2018	37171921	2.41 %	875808	-62920.0	17.4	5.26	57	24.9 %	9273302	0.49 %	7631091040	38
3	Afghanistan	2017	36296113	2.58 %	913081	-62920.0	17.4	5.26	56	24.7 %	8971472	0.48 %	7547858925	39
4	Afghanistan	2016	35383032	2.82 %	969429	-62920.0	17.4	5.26	54	24.5 %	8670939	0.47 %	7464022049	39

	Year	Population	Yearly Change	Migrants (net)	Median Age	Fertility Rate	Density (P/Km²)	World Population	Rank
count	3618.000000	3.618000e+03	3.618000e+03	3.618000e+03	3618.000000	3618.000000	3618.000000	3.618000e+03	3618.000000
mean	1994.166667	2.832801e+07	3.824206e+05	-4.818684e+00	25.474986	3.887186	249.710614	5.694506e+09	101.052792
std	21.724088	1.144600e+08	1.429847e+06	1.060564e+05	8.218262	1.985077	1209.870554	1.736142e+09	58.121232
min	1955.000000	1.898500e+04	-6.730240e+05	-1.077397e+06	14.400000	0.850000	0.000000	2.773020e+09	1.000000
25%	1975.000000	1.056333e+06	6.570250e+03	-1.228375e+04	18.500000	2.090000	19.000000	4.079481e+09	51.000000
50%	1997.500000	5.178470e+06	6.008850e+04	-9.000000e+02	22.750000	3.400000	65.000000	5.943853e+09	101.000000
75%	2016.000000	1.720896e+07	2.693282e+05	5.000000e+03	31.600000	5.740000	149.750000	7.464022e+09	151.000000
max	2020.000000	1.439324e+09	2.067648e+07	1.771991e+06	48.400000	8.800000	21645.000000	7.794799e+09	210.000000

	country	Year	Population	Yearly % Change	Yearly Change	Migrants (net)	Median Age	Fertility Rate	Density (P/Km²)	Urban Pop %	Urban Population	Country's Share of World Pop	World Population	Rank
0	Afghanistan	2020	38928346	2.33 %	886592	-62920.0	18.4	4.56	60	25.4 %	9904337	0.50 %	7794798739	37
1	Afghanistan	2019	38041754	2.34 %	869833	-62920.0	17.4	5.26	58	25.2 %	9582625	0.49 %	7713468100	37
2	Afghanistan	2018	37171921	2.41 %	875808	-62920.0	17.4	5.26	57	24.9 %	9273302	0.49 %	7631091040	38
3	Afghanistan	2017	36296113	2.58 %	913081	-62920.0	17.4	5.26	56	24.7 %	8971472	0.48 %	7547858925	39
4	Afghanistan	2016	35383032	2.82 %	969429	-62920.0	17.4	5.26	54	24.5 %	8670939	0.47 %	7464022049	39
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
3613	Zimbabwe	1975	6293875	3.54 %	200914	-9109.0	15.4	7.40	16	19.3 %	1215331	0.15 %	4079480606	79
3614	Zimbabwe	1970	5289303	3.42 %	163625	-8400.0	15.6	7.40	14	17.0 %	898584	0.14 %	3700437046	79
3615	Zimbabwe	1965	4471177	3.43 %	138899	-3002.0	16.0	7.30	12	14.4 %	644767	0.13 %	3339583597	91
3616	Zimbabwe	1960	3776681	3.28 %	112679	-1501.0	17.2	7.00	10	12.5 %	472478	0.12 %	3034949748	87
3617	Zimbabwe	1955	3213286	3.19 %	93287	-901.0	18.1	6.80	8	11.5 %	371106	0.12 %	2773019936	91

	country	Year	Population	Yearly % Change	Yearly Change	Migrants (net)	Median Age	Fertility Rate	Density (P/Km²)	Urban Pop %	Urban Population	Country's Share of World Pop	World Population	Rank
0	Afghanistan	2020	38928346	2.33 %	886592	-62920.0	18.4	4.56	60	25.4 %	9904337	0.50 %	7794798739	37
1	Afghanistan	2019	38041754	2.34 %	869833	-62920.0	17.4	5.26	58	25.2 %	9582625	0.49 %	7713468100	37
2	Afghanistan	2018	37171921	2.41 %	875808	-62920.0	17.4	5.26	57	24.9 %	9273302	0.49 %	7631091040	38
3	Afghanistan	2017	36296113	2.58 %	913081	-62920.0	17.4	5.26	56	24.7 %	8971472	0.48 %	7547858925	39
4	Afghanistan	2016	35383032	2.82 %	969429	-62920.0	17.4	5.26	54	24.5 %	8670939	0.47 %	7464022049	39
...	...	...	...	...	...	...	...	...	...	...	...	...	...	...
3613	Zimbabwe	1975	6293875	3.54 %	200914	-9109.0	15.4	7.40	16	19.3 %	1215331	0.15 %	4079480606	79
3614	Zimbabwe	1970	5289303	3.42 %	163625	-8400.0	15.6	7.40	14	17.0 %	898584	0.14 %	3700437046	79
3615	Zimbabwe	1965	4471177	3.43 %	138899	-3002.0	16.0	7.30	12	14.4 %	644767	0.13 %	3339583597	91
3616	Zimbabwe	1960	3776681	3.28 %	112679	-1501.0	17.2	7.00	10	12.5 %	472478	0.12 %	3034949748	87
3617	Zimbabwe	1955	3213286	3.19 %	93287	-901.0	18.1	6.80	8	11.5 %	371106	0.12 %	2773019936	91

Correlation Analysis¶

TO VISUALISE THE DATA ON A MAP¶